Hypergraph Clustering Model Based on Fuzzy Frequent Itemsets Applied in Management of Agricultural Land Evaluation

Abstract

There are two shortages in usual methods for agricultural land evaluation: (1) too many manual interferences into the calculating procession, (2) the relatively large differences of partial units are concealed in certain factors. We designed a hyper graph clustering model in this paper based on fuzzy frequent item sets to conduct the evaluation for quality of agricultural land. The database for land units is composed with the fuzzy feature vectors. It executes the mining association rules to the fuzzy item sets given by the definition of evaluation factors, and analyzes the clusters with the HMETIS segmentation method, finally devises the transactions similarity function to cluster the services. With the following checking example in some region in South China, it not only determines the quality rating of each units of agricultural land, but also gives the quality description for each grade.